Automated Clustering of High-dimensional Data with a Feature Weighted Mean Shift Algorithm
This work provides an incremental improvement for researchers and practitioners using mean shift clustering on high-dimensional datasets, by extending its applicability and improving performance.
The paper addresses the challenge of applying mean shift clustering to high-dimensional data, where many features are irrelevant. It introduces a feature-weighted mean shift algorithm that learns feature importance, outperforming conventional mean shift and state-of-the-art methods while maintaining computational simplicity.
Mean shift is a simple interactive procedure that gradually shifts data points towards the mode which denotes the highest density of data points in the region. Mean shift algorithms have been effectively used for data denoising, mode seeking, and finding the number of clusters in a dataset in an automated fashion. However, the merits of mean shift quickly fade away as the data dimensions increase and only a handful of features contain useful information about the cluster structure of the data. We propose a simple yet elegant feature-weighted variant of mean shift to efficiently learn the feature importance and thus, extending the merits of mean shift to high-dimensional data. The resulting algorithm not only outperforms the conventional mean shift clustering procedure but also preserves its computational simplicity. In addition, the proposed method comes with rigorous theoretical convergence guarantees and a convergence rate of at least a cubic order. The efficacy of our proposal is thoroughly assessed through experimental comparison against baseline and state-of-the-art clustering methods on synthetic as well as real-world datasets.